libpayload: usb: dwc2: Always return 'size' transferred bytes for OUT
Seems like our transferred bytes calculation for OUT transfers that span more than one packet had been wrong, and we just got lucky that we never noticed it before. The HCTSIZ.xfersize register field we're reading only counts bytes transferred by the last packet we sent. OUT endpoints cannot have short transfers -- every transfer should either finish all bytes we wanted to send or end in a proper error condition. Therefore, in the absence of an error we can just conclude that all input bytes have been transferred. BRANCH=veyron BUG=chrome-os-partner:35525 TEST=SMSC95xx netboot on Jerry now works. Change-Id: I57349e697c428df6b56e2f6f62e87652ef1e7a94 Signed-off-by: Patrick Georgi <patrick@georgi-clan.de> Original-Commit-Id: 0abee13b6d89dec12c6fff581ece1836393c7703 Original-Change-Id: Id0a127e6919f5786ba05218277705dda1067b8c3 Original-Signed-off-by: Julius Werner <jwerner@chromium.org> Original-Reviewed-on: https://chromium-review.googlesource.com/293956 Original-Reviewed-by: yunzhi li <lyz@rock-chips.com> Original-Reviewed-by: David Hendricks <dhendrix@chromium.org> Reviewed-on: http://review.coreboot.org/11404 Tested-by: build bot (Jenkins) Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This commit is contained in:
parent
4bfa29e947
commit
e8a8a00342
|
@ -283,7 +283,7 @@ dwc2_do_xfer(endpoint_t *ep, int size, int pid, ep_dir_t dir,
|
|||
|
||||
if (ret >= 0) {
|
||||
/* Calculate actual transferred length */
|
||||
transferred = (dir == EPDIR_IN) ? inpkt_length - ret : ret;
|
||||
transferred = (dir == EPDIR_IN) ? inpkt_length - ret : size;
|
||||
|
||||
if (do_copy && (dir == EPDIR_IN))
|
||||
memcpy(data_buf, aligned_buf, transferred);
|
||||
|
|
Loading…
Reference in New Issue